Figure 1. Dynamics and intracellular traffic of membrane receptors. Membrane receptors at the cell surface can be internalised in intracellular compartment (endocytosis). Then, they can be recycled back to the plasma membrane (recycling) or sorted and degraded in lysosomes (degradation).

 

 

 

Figure 2. Interleukin 2 receptors are endocytosed by a clathrin-independent mechanism. . Fluorescent confocal microscopy images of cells producing (Control) or not clathrin (Clathrin-siRNA). Interleukin 2 receptors (IL2R, green) are internalised in endosomes of cells with or without clathrin (lower images). In contrast, transferrin (Tf, red), a marker of the clathrin-dependent pathway, is not internalised in absence of clathrin (upper right image). Each image shows two cells (white arrows).

 

 

 

 

Figure 3. Working model for the link between endocytosis and actin cytoskeleton. A core complex composed of dynamin, cortactin, Arp2/3 complex and F-actin appears to be involved in both clathrin-dependent and -independent pathways. Cortactin is proposed to play a central role in connecting endocytosis to actin dynamics through dynamin. Syndapin, intersectin and mAbp1 are actin-dynamin interacting proteins specifically required in clathrin-coated pits and vesicles formation.In addition, Rac1 and the kinases Paks could regulate the function of cortactin in clathrin-independent uptake.

 

 

 

 

 

Figure 4. Scheme of the infectious cycle of Chlamydia. The whole cycle takes place in the host-cell, in 48 to 72 hours. The infectious form of Chlamydia (EB) differentiates once inside the cell in the proliferative form (RB), which multiplies and differentiates to EB at the end of the cycle.

 

 

 

Figure 5. Epithelial cell infected with Chlamydia. Bacteria (red) are localized inside the inclusion (arrow). The nucleus is visualized in blue (DNA staining) and the cytosol in green.

 

 

 

Figure 6. Model of IncA structure based on the structure of the SNARE complex. One monomer and the backbone of 3 other monomers (yellow ribbons) are shown. The model was done in collaboration with the Unité de Bioinformatique Structurale, Institut Pasteur.

Figure 7. The bacterial protein NUE accumulates in the nucleus. Hela cells were transfected with NUE (green), which is localized in the cell nucleus (blue). In infected cells, NUE is also translocated in the host cytoplasm by a type III dependent process, and gets to the nucleus where it associates with chromatin.

 

Head of the unit
Alice Dautry-Varsat

 

 
 

 
 
1. RECEPTOR DYNAMICS AND INTRACELLULAR TRAFFIC
( A. Grassart, N. Sauvonnet)


Binding of ligands to membrane receptors at the cell surface can be followed by their internalization by endocytosis. Once internalized, receptors and ligands are sorted in intracellular membrane compartments and can be degraded or recycled back to the cell surface (Figure 1).


Endocytosis pathways


Receptor-mediated endocytosis allows cells to communicate with their environment via membrane receptors which bind macromolecules in the extracellular milieu. It is an essential process for cells since it controls many functions including nutriment uptake, growth factor, hormone responses and antigen presentation. It represents also a way of entry of some pathogens into cells.
Receptor-mediated endocytosis through clathrin-coated pits and vesicles has been by far the most thoroughly investigated. Receptor-ligand complexes concentrate in coated pits, which invaginate to form coated vesicles that bud from the plasma membrane and rapidly loose their coat. These vesicles then fuse with intracellular compartments named endosomes (Figure 1).
In order to find other endocytic routes, we have blocked specifically clathrin-mediated endocytosis: by using dominant negative mutants of the protein Eps15, a component of the clathrin-coated pits (22, 29), and by depleting clathrin form the cells by RNA interference (52). This allowed us to identify the interleukin 2 (IL-2) receptor as the first marker endocytosed by a clathrin-independent mechanism (Figure 2). IL-2 receptors are composed of three associated proteins, α, ß  and γc. The ß and γc receptors belong to the cytokine receptor family. The ß receptor is shared by the IL-2 and IL-15 receptors. The γc receptor is particularly important in the immune system since it is shared by receptors for IL-2, 4, 7, 9, 15 and 21. Endocytosis of ß and γc is rapid and efficient. The ß and γc receptors are not found in clathrin-coated structures, instead they are partially recruited into membrane microdomains enriched in cholesterol and sphingolipids, named "rafts". This new endocytosis pathway is regulated by Rho family GTPases and actin polymerisation. We have also shown that the GTPase dynamin, involved in clathrin vesicle scission, is also required for clathrin-independent endocytosis (5, 40, 52). Interestingly, dynamin also binds several proteins that interact with actin and thus could create a link between endocytosis and actin cytoskeleton. We have tested the involvement of these partners in γc receptor uptake and we have shown that cortactin, that binds to dynamin and actin, is also necessary for clathrin-independent endocytosis. Because cortactin interacts with different activators of actin polymerisation, we propose a working hypothesis whereby dynamin, cortactin and F-actin constitute a core complex that would link endocytosis to actin dynamics in both clathrin-dependent and independent endocytosis (Figure 3).

We investigated the regulation of proteins commonly involved in multiple endocytic routes. We found first specific actors of the clathrin-independent uptake. The Rho GTPases Rac1 and its downstream targets, the p21-activated kinases (Pak) 1 and 2 are specifically required for IL-2R entry, in contrast to the clathrin-dependent pathway. Moreover, we found that cortactin was a target of the kinases Pak1 and Pak2, revealing a cascade Rac1-Pak-cortactin which switchs on specifically the clathrin-independent internalization. Therefore, although some proteins, such as cortactin, are required for several entry routes, their regulation is different upon the pathway studied.


Sorting of intracellular receptors : a role for ubiquitin


After internalization, IL2 receptors reach endosomes, from where ß and γc chains are targeted to lysosomes and degraded. We have shown that the IL2 receptor ß chain contains a signal , responsible for its sorting to lysosomes. It is sufficient when added to a chimeric membrane protein to target it to lysosomes, once it is internalized (16, 27).
Interestingly, the ß chain is mono-ubiquitinated ; if its ubiquitination is prevented, its sorting to late endosomes/lysosomes is inhibited while its internalization is not affected (39). Furthermore, the c-Cbl and NEDD4 ubiquitin ligases, and the de-ubiquitination enzyme DUB-2 modify γc ubiquitination, its expression and intracellular routing (53). In conclusion, the ubiquitination and de-ubiquitination machinery is involved in intracellular sorting of these receptors, thus controlling their surface expression.



2. HOST-PATHOGEN INTERACTIONS : THE MODEL OF CHLAMYDIA
(M. Essid, S. Perrinet,  A. Subtil, F. Vromman)


Chlamydiae are bacteria that proliferate only within eukaryotic host cells. The two species pathogenic to humans, Chlamydia trachomatis and Chlamydia pneumoniae, cause a number of diseases, including trachoma, pelvic inflammatory disease, or pneumonia.
Primary infections are often minor or asymptomatic; the sequelae, blindness, sterility or ectopic pregnancy appear long after infection. Throughout their cycle in the host cell, Chlamydia remain in a membrane-bound compartment called an inclusion (Figures 4,5). At the end of the cycle, the host cell is lysed and infectious forms are disseminated. We investigate the interactions between the bacteria and the cells during infection (47, 48).

 

Secretion of bacterial proteins in the membrane of the inclusion


During the Chlamydia development cycle, the volume of the inclusions increases considerably, until they occupy a large portion of the cytoplasm (Figure 5). The inclusion membrane contains lipids that come from the host-cell. It also contains proteins produced by the bacteria that proliferate inside the inclusion. We have shown that Chlamydia use a type III secretion mechanism to translocate proteins into the inclusion membrane and which are named Inc proteins (38). One of these, IncA, associates with the inclusion membrane and is able to oligomerize. We have shown that IncA has biochemical properties similar as SNARE proteins and we modeled IncA tetramers in parallel four helix bundles based on the structure of the SNARE complex, a conserved structure involved in membrane fusion in eukaryotic cells (Figure 6) (50). IncA plays an important role in the recruitment of several SNARE proteins from the host cell around the inclusion. SNARE-like motifs which we identified in IncA are necessary for this function. Our work demonstrated for the first time mimicry of a fundamental motif of the eukaryotic cell, the SNARE motif, by a bacterium (58, 61).

 

Chlamydia use type III secretion to secrete more than 10% of the proteome into the host

We have searched for type III secretion signals in proteins of unknown function coded by different sequenced Chlamydia genomes. This systematic approach led to the identification of more than twenty proteins that are candidates to be secreted during infection. Added to the 60 to 100 Inc proteins encoded by each Chlamydia genome, these numbers show that the bacteria use a large proportion of their genome to encode proteins that are active not in the bacteria, but in the host (55). These proteins are very likely important for Chlamydia pathogenicity and their function is under study.

Bacterial proteins targeting the host nucleus

NUE is a novel type III secreted proteins we recently identified. After translocation through the inclusion membrane, the protein accumulates in the host nucleus. NUE associates with chromatin during infection. We have shown that NUE has histone methyl transferase activity in vitro, indicating that this activity might result in chromatin modification during infection.

Chlamydia pneumoniae and atherosclerosis : clinical studies

There is growing evidence that chronic infection by C. pneumoniae may contribute to the development of atherosclerosis. We have developed a method to detect bacteria on balloons used for coronary angioplasty (56). This new tool will permit exploration of the possible link between infection and early stages of atherosclerosis. It could help in selecting and following patients for therapeutic trials.